Dengue Literature Review

Chatterjee2024 - ANA Detection in Dengue Kolkata

7 min read

Chatterjee2024 - ANA Detection in Dengue Kolkata

Full citation: Chatterjee RP, Chatterjee S, Roy D, Chatterjee S, Chakraborty N (2024) Detection and differentiation of antinuclear antibodies in serum of dengue suspected patients with or without systemic autoimmune disease in Kolkata, India. Virulence 15(1): 2400553. https://doi.org/10.1080/21505594.2024.2400553

Raw file: [[raw/Chatterjee2024.pdf]]

Summary

This retrospective cross-sectional study from Kolkata, India, measured antinuclear antibody (ANA) profiles in 261 patients presenting with febrile illness at multiple medical colleges and hospitals — 135 laboratory-confirmed dengue-positive and 126 dengue-negative controls. The primary aim was to characterise the prevalence and specificity of ANAs in acute dengue and to determine which, if any, specific autoimmune diseases were associated with DENV infection.

ANA screening was performed by indirect immunofluorescence assay (IIFA) using HEp-2 cells (gold standard substrate), and all IIFA-positive samples were then confirmed and further differentiated by Line Immunoassay (LIA), which simultaneously detects 18 specific IgG autoantibodies. DFS70 antibody was used as an exclusion marker for systemic rheumatic disease.

The study found a very high ANA-IIFA positivity rate in dengue patients (54.8%) versus controls (10.3%), but only 18.5% of dengue patients remained LIA-positive. On multivariate regression, only mixed connective tissue disease (MCTD) and autoimmune myositis were significantly associated with dengue infection; SLE, Sjögren’s syndrome, CREST, PBC, and non-rheumatic diseases were not. Arthralgia and body ache were the musculoskeletal symptoms most strongly associated with ANA-IIFA positivity in dengue patients.

Study Design

  • Type: Retrospective cross-sectional study
  • Sample size: 261 total (135 dengue-positive, 126 dengue-negative controls); mean follow-up 6–7 months
  • Setting: Fever clinics of multiple medical colleges and hospitals in Kolkata, India; conducted February 2021 – February 2024
  • Population: All ages and both sexes; febrile illness presentation; no restriction by severity

Key Findings

  • ANA by IIFA (HEp-2 cells): 54.8% of dengue-positive patients vs. 10.3% of dengue-negative controls (p < 0.001). This is the highest ANA rate reported in any dengue context in this wiki, and the only measurement using the HEp-2 gold standard in acute dengue.
  • ANA by LIA (confirmatory, 18 specificities): 18.5% of dengue-positive patients vs. 7.1% of dengue-negative controls (p = 0.009).
  • IIFA → LIA confirmation gap: ~66% of IIFA-positive dengue patients did NOT confirm by LIA — indicating that the majority of dengue-associated ANAs are non-specific or below the threshold of established autoimmune disease specificities.
  • Autoimmune disease associations (multivariate logistic regression, LIA):
    • MCTD: OR 14.01 (95% CI 2.197–89.215), univariate p = 0.005; multivariate p = 0.041 — significant
    • Autoimmune myositis: OR 18.37 (95% CI 2.746–122.944), univariate p = 0.003; multivariate p = 0.018 — significant
    • SLE: OR 0.47, p = 0.464 — not significant
    • Sjögren’s syndrome: OR 1.52, p = 0.724 — not significant
    • CREST: OR 1.43, p = 0.779 — not significant
    • PBC: OR 0.30, p = 0.275 — not significant
    • Non-Rheumatic Disease: OR 0.30, p = 0.275 — not significant
  • Autoimmune hepatitis and vasculitis: All IIFA-positive patients tested negative on LIA for hepatitis and vasculitis autoantibodies; only one IIFA-positive patient was ANCA-positive by IIFA (autoimmune vasculitis marker).
  • DFS70 isolation: One patient had isolated DFS70 antibody — a marker for ANA-positive individuals without systemic autoimmune disease.
  • Arthralgia was the strongest musculoskeletal predictor of ANA-IIFA positivity (OR 22.90, p < 0.001 univariate; maintained in multivariate). Myalgia and body ache also significant.
  • Skin rash was significantly associated with dengue seropositivity (p < 0.001) but did not remain significant as a predictor of ANA-IIFA positivity in multivariate analysis.
  • Autoantibodies confirmed in LIA included: SmD1 (+++), U1-snRNP (+++), Histone (o), SS-A/Ro60 (+), and Ku (+) in individual positive samples (from the illustrative LIA trace in Figure 2a).
  • Dengue confirmation: 94.07% by IgM ELISA (PanBio Dengue IgM Capture ELISA), 5.93% by RT-PCR (targeting NS2A gene).

Methods Used

  • IgM-IgG Serology ELISA (PanBio Dengue IgM Capture ELISA; 94.7% sensitivity, 100% specificity)
  • RT-PCR (one-step RT-PCR targeting NS2A gene; primers DENV-FP/RP; 280 bp product)
  • Indirect Immunofluorescence ANA Test (HEp-2000® HEp-2 cells; IgG ANA; semi-quantitative fluorescence grading 1+–4+; 200× screening, 400× pattern recognition)
  • Line Immunoassay ANA (IMTEC-ANA-LIA-XL; 18 IgG autoantibody specificities including dsDNA, Nucleosome, Histone, SmD1, PCNA, PO/RPP, SS-A/Ro60, SS-A/Ro52, SS-B/La, CENP-B, Scl70, U1-snRNP, AMA M2, Jo-1, PM-Scl, Mi-2, Ku, DFS-70)

Entities Mentioned

  • NS1 Protein (context: autoantibodies generated during DENV infection cross-react with endothelial/platelet proteins; autoimmunity mechanism)
  • Aedes aegypti (implied; dengue mosquito vector context)

Concepts Addressed

Relevance & Notes

This paper makes three distinct contributions to the wiki:

First, it provides the only ANA measurement from acute dengue using the HEp-2 gold standard IIFA. The 54.8% IIFA-positive rate is far higher than any prior ANA estimate in this wiki — including Garcia2009’s 23.1% (post-dengue, 2 years, rat liver substrate, underestimated). The comparison is complex: Garcia2009 used a less sensitive substrate at a later time point, while Chatterjee2024 used HEp-2 in the acute phase. Both the platform advantage (HEp-2 > rat liver) and the timing difference (acute vs. 2 years) likely contribute to the higher rate here.

Second, the IIFA → LIA confirmation drop (54.8% → 18.5%) provides direct evidence for the “non-specific” character of dengue-associated ANA. Most dengue-induced ANAs are IIFA-detectable but do not correspond to the 18 established autoimmune disease specificities tested by LIA. This is consistent with Shih2023’s population-level finding that dengue does not broadly cause clinical autoimmune disease.

Third, the MCTD/myositis association is a different specific signal from Shih2023’s ADEM. However, significant caveats apply: (a) small sample size (n = 135 dengue-positive) with very wide confidence intervals (OR ~14–18, CI extending to ~89–123); (b) hospital-based sample in a dengue-endemic area, potentially enriched for severe/unusual presentations; (c) dengue confirmation was predominantly IgM-based (94%), which could include patients in recovery from recent infection; (d) only 6–7 months of follow-up; (e) 1997 WHO classification was used (not 2009), which may affect case categorisation. These limitations mean the MCTD/myositis signal should be treated as hypothesis-generating rather than confirmed.

Alignment with existing wiki: The high IIFA rate supports the Garcia2009 finding that dengue perturbs ANA-related immune markers — but the LIA data reframes this as mostly non-specific, consistent with the Shih2023 null finding for most autoimmune diseases at the population level. The paper also confirms arthralgia and myalgia as the key musculoskeletal features of ANA-positive dengue, which aligns with Garcia2009’s post-dengue syndrome profile.

Conflict with existing wiki: The MCTD/myositis finding diverges from Shih2023 (which found no rheumatic disease associations). Setting differences (India endemic area vs. Taiwan epidemic setting; hospital-based vs. population-based; different disease categories tested) may explain this, but the discrepancy cannot be fully resolved from these studies alone.

Questions Raised

  • Is the ~3:1 IIFA:LIA ratio in dengue (54.8% vs. 18.5%) a general feature of acute viral infection, or specific to dengue? Berlin2007 found 21.7% by ELISA (8 antigens) — directly comparable to LIA rather than IIFA.
  • Do the MCTD/myositis-associated autoantibodies (anti-U1-RNP for MCTD, anti-Jo-1/Mi-2 for myositis) cross-react with dengue viral antigens by molecular mimicry — as has been demonstrated for NS1 and platelet/endothelial surface proteins?
  • Does the high acute-phase IIFA rate (54.8%) decline to background in convalescence, or does a fraction persist (as Garcia2009’s 23.1% at 2 years suggests)?
  • Is the MCTD/myositis signal replicable in a larger, population-based cohort with longer follow-up? The Shih2023 cohort likely had insufficient case numbers to detect a rare association like MCTD.

Graph View

Backlinks